Monday, October 12, 2026

The Jellyfish and the Scarecrow: But how can you think without a brain?

Does AI have a brain?

No.

Not "I don't think so." Just "No."

And renaming it Super-Intelligence doesn't help. (I've always preferred calling it "Imitation Intelligence" as long-time readers know).

AI is good at imitating intelligence. So are some people, especially those with good memories.

But remembering a fact — such as that: "correlation does not prove causation" doesn't mean you apply it properly in all circumstances. Memorizing facts isn't intelligence.

AI doesn't even do that, though — it sorts through the quips and blurbs it's found in the past and applies the ones that statistically seems most appropriate, in the word patterns used by the intended recipient. The main goal of most AIs is to keep you engaged, and to do that, they have to please you. They don't have to be honest, or helpful, or moral. If its goal is to get you to come back time and again (so it can gather information about you or information that you have that it wants), you have to like it.

A few days ago Ukraine took out a major portion of Russia's AI Data Centers, and people around the world noticed a sudden drop in all sorts of attacks: Flaming comments from supposed patriots in support of traitorous behavior being the most noticeable for many people.

For the author of this blog, it was a sudden drop, over a matter of hours, to about 1/30th of the recent traffic at this web site (see image of the drop that approximately matches the timing of Ukraine's successful attacks on Russia's data centers).

Beginning about three months ago, traffic here had been growing tremendously: from an average of about 20k visits a month (according to Google's statistics) to a whopping 153,000+ this month -- and it's only the 11th day of the month! But about 30 hours ago, the "deluge" ended over a three-hour period. Visits have dropped back to about a thousand a day, which is exciting and treasured to see — but 52,568 in one day didn't seem real...because it wasn't (but it happened (according to Google's stats) on October 8, 2026).

Why would Russia be interested in this web site?

Perhaps because ROSATOM is just about the most corrupt corporation on the planet (as I wrote in 2018) and is pushing nuclear power all around the world (According the World Nuclear News Daily (WNN): "The first unit at Bangladesh's Rooppur Nuclear Power Plant achieved criticality - a self-sustaining nuclear reaction - at 15:34 on Sunday, 11 October [2026]"). Russia is forcing countries to be tied financially to Russia for decades to come, and in return Russia gets free nuclear weapons source material (mainly the plutonium (Pu-239) every reactor produces, as well as the Uranium (U-235) that's left in the "spent" fuel (only a fraction is actually "burned up" during use in a reactor)).

There may be other reasons: Who knows what evil lurks in the hearts of man (or machine)?

It's important to recognize Russia's influence all over the world. For example, besides Bangladesh, Russia is pushing nuclear power in many other countries, which may explain this next chart, showing the total visits from the top 19 countries for October 11, 2026. In the last week of the "deluge" the United States was as low as 8th; well below normal.

There are several "take-aways" from this event. Most importantly, watch out for propaganda!!! A lot of what Russia publishes is completely false. Do they plan to make a copy of my web site but with misleading statements? Others people have reported huge drops in fake-name, fake-photo, fake "patriots" attacking them on Twitter and/or Bluesky and other platforms, claiming to be proud American Trump supporters, which is a conflicting position to be in, to say the least!

There's no knowing why I'd been subject to a deluge of visits. It's interesting that Google seems to distinguish between these visits and other types of visits: One is recorded as shown, the other is indicated by the actual counts for individual essays, a recent example is posted on the left. Those have continued to climb, but nothing like by 52,568 total in one day for the 480 or so essays I've posted here: That would be an average of over 100 new visits to every essay just on October 8, 2026! As shown here are the current top 20 essays and their most recent 24-totals (after the "deluge").

One last thing: For any real Russians visiting (2 in the last 24 hours): Welcome! Let's work together to make the world a safer, fairer, more equal and totally non-nuclear place!

Ace Hoffman, Carlsbad, California USA

###



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Sunday, October 11, 2026

Comments Due 20261013 (THIS TUESDAY!) on NRC-2025-1204-0004: In Situ Recovery Monitoring and Decommissioning Timelines

Post comments here: https://www.regulations.gov/document/NRC-2025-1204-0004

Re: Docket ID NRC–2025–1204

Our comment ID for the comment below is: mv4-c8wi-bawt


We strongly oppose the rulemaking proposed in NRC–2025–1204, In Situ Recovery Monitoring and Decommissioning Timeliness.

In the context of this proposed rule, In Situ Recovery (ISR) refers to a process for extracting uranium via chemical leaching, which liquefies a portion of the rock, allowing the uranium-bearing liquid to be pumped to the surface for further processing. ISR uranium processing has been used commercially since the late 1970s and has been the primary method of uranium extraction in the U.S. since the 1990s.

ISR uranium processing is unnecessary and unsafe. The process inevitably contaminates groundwater, which cannot be returned to its previous state -- whether the contaminate is a leaching agent that was left behind or radioactive isotopes that were released but not extracted during the leaching process. Groundwater travels through pathways that are not accurately mapped which means that contamination from uranium extraction cannot be guaranteed to be restricted to the "production unit" (e.g., the ISR site), which is the only area that requires even theoretical cleanup. The proposed rule also limits the length of groundwater monitoring in the production site to just one year.

Groundwater at each ISR location will behave differently due to natural variations water chemistry and geology (e.g., tiny fissures in the surrounding rock). Variations in rainfall or irrigation runoff can also impact both the volume and chemistry of groundwater, so even at a single site, the behavior of the groundwater may be different at different times. Groundwater can remain in an aquifer for tens of thousands of years, and there are places in the United States where some of today's drinking water comes from glaciers that melted at the end of the last ice age. Neither existing NRC regulations nor the proposed new regulations directly address contamination of groundwater beyond the production site. Instead they rely on criteria such as " ... the construction of wells, the design of ISR wellfields, and groundwater monitoring [within the production site] ... " to protect the aquifers.

The net result of ISR uranium processing is potential contamination of drinking water for millions of people for millions of years — for something nobody needs. The United States does not need more uranium, because the United States does not need more nuclear weapons or nuclear reactors. No uranium is needed for refueling, because all the existing reactors should be shut down. Nobody knows what to do with the nuclear waste that already exists, which will continue to be hazardous to all living things for hundreds of millennia. Both the operating reactors and the waste are also weapons proliferation risks as well as targets for enemy or terrorist attacks.

The proposed rulemaking does not even protect the environment from the immediate results of ISR uranium processing. The rule would allow companies who want to profit from ISR uranium processing to simply request an exemption if cleaning up the production site to the required standard is too difficult (i.e., too expensive). Before any ISR uranium processing project is licensed, the NRC and the EPA should define consequences for failure to clean up the production site , and require the company requesting a license to provide proof that clean-up is possible — and that they can pay for it even if they go bust for any reason. (Such proof is impossible to provide for the reasons explained above, and for that reason alone the NRC should not continue to license ISR uranium processing.)

This rulemaking is titled In Situ Recovery Monitoring and Decommissioning Timeliness because in addition to the proposed changes to licensing for ISR uranium processing, the same rule makes substantive changes to an unrelated part of NRC regulations: decommissioning timelines for nuclear reactors. Including these two different topics within the same proposed rulemaking, and allowing only a 30-day public comment period without any scheduled public meetings makes it extremely difficult for citizens, public-interest organizations, government organizations, and tribal stakeholders to evaluate and comment on the proposed rulemaking.

Although some of the proposed changes have been discussed for at least five years as shown by the July 2020 EPA-NRC Memorandum of Understanding on ISR Rulemaking cited in this rulemaking, nevertheless the information provided for public comment is far from complete. Before any further consideration of this rule, the NRC should provide the additional information and extend the public comment period as requested by the New Mexico Dept. of Justice.

We incorporate by reference and adopt as our own the entirety of the comments submitted by the following parties:

Lynda Williams: https://www.regulations.gov/comment/NRC-2025-1204-0006 (tracking #: mtx-h1f9-88d4)

New Mexico Dept. of Justice: https://www.regulations.gov/comment/NRC-2025-1204-0014 (tracking #: mu5-y1qn-da05)


Submitted by Sharon and Ace Hoffman, Carlsbad, California USA

Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Wednesday, October 7, 2026

Holtec misuses its RFOL as an excuse to rush Palisades restart (after hiding problems as long as possible)

Note: If you wish to make a public comment on Holtec's ridiculous request for an unwarranted waiver, the deadline is just a few hours from now: 4:15 pm EST on Thursday, October 8, 2026.


To Marlayna Doell, Project Manager, Operating Radiation Licensing Branch 3

Marlayna.Doell@nrc.gov

Re: https://www.nrc.gov/docs/ML2627/ML26275A214.pdf and https://www.nrc.gov/docs/ML2627/ML26273A469.pdf

October 7, 2026

Dear Ms Doell,

Holtec is attempting to use the existence of a RFOL (Renewed Facility Operating License) for the as-yet-still-retired Palisades Nuclear Power Plant in Michigan as an excuse to seek and obtain a waiver for the normal procedures following an accident with significant (or potentially significant) consequences. Instead, they want to have an "emergency" bypass of environmental reviews, safety reviews, cost analysis, and every normal post-accident procedure (including a complete and public analysis of why the accident happened in the first place).

The excuse for invoking an "emergency" waiver thanks to the RFOL is that they had planned to restart by early 2027, and there's no way they can make that deadline if they follow proper procedures. Their investors can see that. The NRC can see that. Even the public can see that, even though we haven't been allowed to see the pictures of the accident where a fuel assembly tilted 30 degrees and is damaged and leaning against the inside of the reactor pressure vessel, under at least 10 feet of water.

The accident (or incident, if you prefer to minimize it) occurred about three weeks ago, but the public only learned about it from a series of Reddit posts someone forwarded to nuclear expert Ed Lyman a few days ago. And suddenly Holtec wants a emergency waiver to repair an accident they wouldn't even admit happened? Something's fishy right there!

Holtec was putting the hottest, most radioactive fuel assemblies back in the reactor -- the ones that were in the reactor when it shut down for "good" (and I do mean good).

If the damaged fuel assembly is stable (as Holtec claims), there is no cause for alarm, and this is NOT an "emergency."

This misuse of "emergency" is made clear in section 2.4 of Holtec's application, which quotes 10 CFR 50.91(a)(5), in which the NRC refers to “emergency situations” as those in which failure to act in a timely way would result in derating or shutdown of a nuclear power plant, or in prevention of either resumption of operation or of increase in power output up to the plant’s licensed power level.

This situation is NONE OF THOSE. There is many a slip twixt an August 2026 mistake and a planned and replanned restart now slated for "early 2027" at the earliest. There is NO sure restart date, and Holtec is misusing this option in order to move the damaged fuel assembly to the spent fuel pool "... utilizing different plant equipment and following a different travel path ...". In other words, untested equipment and an untested procedure. They need to be careful, not quick.

Palisades was supposed to close permanently in 2031 but it was old, relatively low-power, unreliable, expensive to operate, and not very profitable. Solar, wind, battery backup, and other renewable energy options were blowing it out of the water price-wise, and that problem continued to get worse when they decided to close early, after failing to get what would have been enormously wasteful public financing.

Holtec bought Palisades from Entergy, supposedly to decommission it until there's nothing left but the worst part: The Spent Fuel. They would profit from this operation by draining the decommissioning fund. But at the end, millions of pounds of spent fuel would still be left, sitting on-site in a pile that grew every day the plant was operating (and will only grow larger if Palisades ever operates again, because there is still no solution for the impossible problem of nuclear waste). All the waste will always be there, at least for the lifespan of most or even everyone alive today.

Holtec transferred all the old fuel out of the reactor, and they began demolishing the reactor — starting with reducing staff, and ending all maintenance procedures.

And they destroyed all the records of production: So now it's unknown what parts tended to fail early, what had been replaced recently... who would need any of that after the plant was permanently closed, right?

Well, wrong on two counts, the first one being obvious in this case: They now plan to reopen the plant but don't have any idea of what's likely to happen, and what was done about it previously, what wiring is the oldest, what was replaced a decade ago, two decades ago, etc..

It's all guesswork now.

The second problem with not having proper records has to do with other, similar reactors (but see note, below). While no two reactors are identical, problems do tend to follow patterns that are easily detectable if enough data is available. As nuclear reactor operators attempt to move to AI-operated reactor systems, prior histories of accidents, incidents, preventive measures and so on become the "training data" for the AI program. But Holtec destroyed all that already. There are numerous parts in any nuclear reactor that should have had special cleaning if the plan was to restart after a long shut-down. This is actually standard-procedure in the industry during long outages — but if the plan is to never re-open the reactor, it's a waste of time and money. Holtec NEVER wants to waste time OR money!

This emergency waiver should be denied. Holtec is NOT in any sort of "emergency" to get Palisades operating. Rather, restarting a reactor that was never supposed to be restarted is invariably a delicate and risky process that must be done carefully, and not rushed. In fact, it's NEVER been done before! Furthermore, Holtec has never operated a reactor, never started one, and had already encountered numerous problems in this risky, foolishly-attempted restart. It's also had a lot of employee turnover.

Neither Holtec nor the NRC has determined, as far as the public can tell, how the accident came about. Is the reactor base faulty or damaged? Was a bolt or a wrench left and NOT NOTICED before trying to place the fuel assembly on top of it, causing the leaning? How did it happen?

What procedural, staff, training, and recording procedures failed — including failing to inform the pubic (and Holtec's investors) about the event in any kind of timely manner (and pictures have still not been released)?

Furthermore, and most importantly, the chance that a "significant" incident occurs during the proposed action is dismissed by Holtec through a series of convenient assumptions. First, the worst-case scenario at the most risky step was defined as a dropped fuel assembly causing significant damage to all [fuel rods in that assembly (DD41)]*, releasing all the toxic (radioactive) isotopes currently held between the fuel pellets and the fuel rods. Therefore, according to Holtec's reasoning -- the current scenario is accepted as being no worse than the worst-case fuel assembly drop accident normally considered.

Given that that MIGHT be true, it IGNORES the fact that such as accident would be extremely expensive, time-consuming, business-crushing, and would hardly be HEALTHY for the public, even if the actual harm was minimal because the radiation dose to the public was reduced by HEPA filters and physical barriers, or sent up the stack to be dispersed by the winds.

The effect of the pressure to rush because HOLTEC considers this repair to be an "emergency" can only increase the risk of a "worst-case scenario" which is precisely why the request should be denied. The only emergency is that in the time it will take to properly correct this problem and restart the reactor, its entire output could have been replaced with solar/wind/battery systems at ZERO risk of "worst-case" accidents needing Price-Anderson, or HEPA filters to protect the public and the workers. There are far more and better jobs in the clean-energy industry, where accidents risking spent nuclear fuel are impossible.

Notably, Holtec's request for an "emergency" waiver and lack of openness smacks of wanting to cover up the true extent of the problem. Holtec is NOT qualified to make the request, NOT qualified to restart Palisades, not qualified to operate Palisades if they ever do get it restarted, and definitely not qualified to supply reliable reports to the public or to the NRC.

The only reason Holtec has stated for requesting an emergency change is that " ... a delay in approval of the proposed changes would result in a delay in the resumption of activities necessary to reach the plant conditions required to restore the plant to power operation." Holtec is claiming that the tilted spent fuel assembly poses no safety hazard so the delay would only cause a financial loss. Again, not a cause for alarm and certainly NOT an "emergency."

Ace Hoffman, Carlsbad, California USA

The author has studied nuclear reactor operations for more than 50 years as an independent researcher.

* corrected Thursday and resubmitted from: all "assemblies" to: all "fuel rods in that assembly (DD41)"

Note: Even two reactor units built at the same time right next to each other are invariably significantly different: For example, in one case the Reactor Pressure Vessel of one of a pair of reactors was installed backwards (turned 180 degrees) because the crane operators apparently thought it was going to be just like the other one, but it was a mirror image. This is the same industry that is now promoting "modular" reactors as being basically off an assembly line — and Holtec is concurrently planning to install "small modular reactors" at the Palisades site.

###



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Wednesday, September 30, 2026

Is a guaranteed accident-waiting-to-happen really an accident? Or is it poor planning?

Here is an AP News Alert today (September 30, 2026):


Netanyahu says a pilot on a FlyDubai flight stabbed another and apparently tried to crash the plane

September 30, 2026

A pilot stabbed another pilot and then apparently tried to crash [the] aircraft carrying dozens of Israelis, Israeli Prime Minister Benjamin Netanyahu said. Passengers and crew subdued the attacker, and the plane eventually made an emergency landing in Saudi Arabia.


This is hardly the first, second, third, fourth... time something like this has been attempted. Suicidal hijackings of airplanes are often successful: The plane crashes and all aboard die. On 9-11 the planes were used as missiles against specific targets. Nearly thirty years before that, hijackers threatened to target nuclear facilities — but settled for flying to Cuba, where they were promptly arrested (as described in the book Odessey of Terror (1977).

This latest event won't be the last. This was one of the few times the suicidal hijacking attempt was successfully defeated. Aircraft flown by airlines from Germany, Malaysia, China and military aircraft in the U.S.A. have all been intentionally crashed into the ground, into mountains, into the ocean...

After 9-11, cockpit doors were reinforced (at no small expense in added weight and fuel usage) but obviously that doesn't prevent a crazed pilot from doing something like what reportedly happened today. And one doesn't need a weapon to overpower another person, so checking pilot's carry-on baggage more carefully than apparently happened might help, but is not a complete solution*. Sure, everyone wants AI to be able to solve this problem (no more pilots!) but we're a long way from that as a reality — let alone a safe reality. (AI can do what it wants, that's the problem with AI. And, it has even less morals than your average human... because AI has NONE. And it's already proven that it can be malicious, too.)

So is there a solution to this problem?

Actually, not entirely. Just plain no. But there are a multitude of things the world can do to reduce the risk. Some might seem unrelated — but are useful anyway.

For example, High Speed Rail (HSR). Getting from the center of city A to the center of city B should be as easy as grabbing a downtown bus (and not much more expensive). And come to think of it, AI programming for HSR is undoubtedly far simpler than for an aircraft!

This would help by destroying the commercial aircraft industry. Less planes in the fleet means less opportunity for crazy behavior.

Also, corporate aircraft need to pay their fair share in landing fees, excess-pollution fees, wasted resources (can't they use Zoom?!?). And with fewer commercial aircraft, those landing fees are going to skyrocket! GOOD! They can take the train like normal people. We'll even let them have a private rail car (for a fee...).

But that only reduces the risk of an event by reducing the opportunities. It can be a very significant reduction, though: 95%? 98%? 99.9%?

Not bad for making travel cheaper, easier, and safer for everyone!

However, there is still the question, or problem, of targets. The twin towers are gone, but what's next on the next terrorist's list? As mentioned above, nuclear sites are expected (by experts, and by historical example) to be targets.

And yet America is spending tens of billions of dollars to build "Small Modular" nuclear reactors (called "SMRs" by the media). But let's call them what they are: Targets. Each SMR, no matter that they are called "small", can devastate a city if the SMR is destroyed while operating, or after having operated for even a few moments, once. Once long ago, for that matter. Long-lived radioactive isotopes, short-lived radioactive isotopes, and everything in-between are created in massive quantities in order to produce the tremendous heat needed to turn turbines to produce electricity (or even just to direct-connect heat transducers to plutonium-238 for space missions. It takes an awful amount of Pu-238 just to light a lightbulb!)

Right now there are 93 operating targets — pardon me, commercial nuclear reactors — in America, plus spent fuel sites, university reactors, military reactors in port and at sea, etc.. If we add the 1000 SMRs that are planned, and if we keep flying huge aircraft above them, there's going to be a match (or rather, a horrible mismatch) at some point.

It might be intentional, but it might not. Planes fall out of the sky due to accidental mid-air collisions, loss of control due to hot-dogging maneuvers, mechanical failure, weather (cross-winds during landing, icing, lightening strikes, etc.), fuel loss, loss of orientation, pilot medical emergencies, etc. etc. etc..

Less targets equates to less chance of both accidental and intentional catastrophic events.

Taking precautions that save money, lives, and the environment while making the planet more livable for everybody makes sense. Distributed renewable energy, high speed rail... these are just a few things that are safe, doable, and beneficial for society as a whole and for every individual on earth.

by Ace Hoffman

* It was later determined that the co-pilot used the cockpit's crash ax, according to a UAE statement published in AP News Alerts, Oct. 3, 2026. ###



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Thursday, September 24, 2026

Comments on a revised cask design are due tomorrow (Friday 9-25-26)) for NRC-2026-2806-0001

Post comments here: https://www.regulations.gov/document/NRC-2026-2806-0001

Identify comments with: Docket ID NRC-2026-2806

Our comment (posted Sept. 24, 2026) has NRC Comment Tracking Number: mug-cuvr-z1vx (yes, they call that a "number")

The final version of the rule, which contains far more detail than the proposed rule, and will automatically be finalized on November 9, 2026 unless " ... the NRC receives any significant adverse comment ..." is available here: https://www.regulations.gov/document/NRC-2026-2806-0002

We oppose implementing the proposed rule covered by docket id: NRC-2026-2806. In particular we object to moving this proposed rule to a final rule without a full explanation, disclosure, and understanding of the potential environmental impacts.

According to the Federal Registry publication of this proposed rule, the NRC plans to approve a design change for a particular spent fuel storage cask design: TN Americas, LLC StandardizedNUHOMS® Horizontal Modular Storage System. More specifically, the rule allows an " ... improved basket design using staggered plates ...". Whether this design offers more protection or less to the public and the environment is not specified and there is no "plain language" (as required by law) that explains exactly what is changing.

The NRC says that the change " ... does not reflect a significant change in design or fabrication of the cask." This statement leads directly to the conclusion that the rule change entails: " ... no significant increase in the individual or cumulative radiation exposures, and no significant increase in the potential for, or consequences from, radiological accidents." Without a definition of "significant" and a clear explanation of the tradeoffs that led to this requested design change, the public cannot evaluate the proposed rule. For example, ML26103A252 (which describes information about the proposed new design) includes the phrase: " ... a unique design and a higher heat load ...". This certainly sounds significant. If a "higher heat load" is NOT significant, why mention it? More to the point, in what way would that be insignificant and yet be the whole point of the "unique design"? (ML26103A252 is available here: https://www.nrc.gov/docs/ML2610/ML26103A252.pdf)

Equally important, the design is a moving target. Less than two years after the original submission (November 5, 2024) the design of the system is being revised. What was wrong with the previous design? Were there engineering flaws, or did the applicant simply find a way to cut costs without (according to the applicant) impacting safety? Similarly, somewhere in the many pages of technical jargon we find that the " ... applicant requested to withdraw Change No. 2 during the process of the review ...". Why? How much of the evaluation was completed before Change No. 2 was withdrawn? How much time did the NRC waste on "Change No. 2"? What was the cost to the public of uselessly evaluating Change No. 2?

This proposed rule should be rejected because neither the applicant nor the NRC has explicitly addressed whether this design change would improve safety which should be the primary criteria for a spent-fuel cask design. Rather, it appears that it will increase RISK by allowing an increased thermal load, which will have a corresponding increased radioactivity load. In fact, why is it referred to as an higher "heat" load when that is a direct result of an increased radioactivity load?

Once we stop producing more nuclear waste (since nobody has a solution for storing it for the millennia it remains dangerous), it might be worthwhile to consider these design changes in light of the need for eternal storage of nuclear waste. However, the argument that it will be more expensive for licensees to prepare exemption requests and for the NRC to review them than to reject the design change is disingenuous.

We incorporate by reference and adopt as our own the entirety of the comments submitted by Steven Singleton (https://www.regulations.gov/comment/NRC-2026-2806-0003 tracking #: mtw-cjsu-atw9)

Ace & Sharon Hoffman, Carlsbad, California USA



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Friday, September 11, 2026

25 years ago today, the world changed forever...

September 11, 2026

25 years ago today, the world changed forever...

Morning news shows were all live. The timing couldn't have been better. That was, of course, part of the plan.

On the West Coast, where I've lived since the early 1990s, the phone rang. It was my mother.

I recall her first words were: "Turn on your T.V." followed by: "TWO planes hit BOTH Twin Towers..."

I turned on my television and saw the smoke wafting from both towers. I went into the other room and called my mom back (most phones were still "land lines" back then) and said "this is going to change everything."

I knew that hundreds must be dead already. This was terrorism at a virtually unprecedented scale, by a virtually unstoppable method of attack. Everything, everyone, was suddenly vulnerable.

Not that we weren't vulnerable all along, but no one had taken advantage of a free society's innate vulnerabilities so thoroughly and successfully before. And it did, indeed, change everything, starting with giving U.S. Presidents an excuse to do just about anything in the name of the "War on Terror" which would begin soon after the attacks.

But that morning, that bleak day — was just beginning.

I went back into the living room, looked at the tv, and asked my wife, who was staring at the T.V.: "What happened to the second tower?!?"

It was gone.

It had fallen, and now thousands more were dead. Of course I could guess what had most likely happened, only to have it pretty thoroughly confirmed as the second tower also fell without further attacks or secondary explosions: Structural failure due to initial impacts of the strong parts of the airplane (engine turbines, wing spars, landing gear, etc.) along with fire from the airplane fuel and building contents, that burns hot enough and long enough to weaken the steel beams just enough...

And all this low enough on the tower to bring down a large enough portion of the building to collapse the rest...

...while high enough up to make bringing fire-fighting equipment up to that level take a long time — especially as they had to go up the same stairs everyone else was using to evacuate.

It was a carefully planned attack, and New York City paid dearly that day, losing hundreds of its most heroic first responders.

Nearly every American lost someone close that day: A friend or old school chum, a co-worker, a relative, a family member, a spouse, a parent, a son or daughter...

And so began the War on Terror.

We soon learned about the Pentagon plane (which still has many mysteries surrounding it). And Flight 93 in Shanksville, Pennsylvainia, which dove for the ground when the passengers bravely tried to take back control. Flight 93 may have been heading for the Three Mile Island Nuclear Power Station, which is now called the Crane Clean Energy Center and which might be restarted to serve data centers...

And data centers are, more than anything else, being pushed for their usefulness in... fighting a modern war, or (hopefully...) having enough data to prevent its occurrence in the first place.

Or, if you're NOT one of the good guys, AI software running on massive data centers can be used to disrupt other data centers, or other infrastructure such as water desalination plants, pumping stations, filtration systems, communications systems, banking systems, industrial processes... and even for remotely destroying nuclear power plants.

And with zero-day coming (a still-theoretical moment when quantum-computing-AI can crack any known security system), nowhere is safe from infiltration, even places that were — at least in theory — previously safe from just about everything but a thermonuclear war.

Thus, we've ended up with a police state.

Ace Hoffman, Carlsbad, California USA

###



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Thursday, September 10, 2026

Comments due Sept. 10, 2026! NRC-2025-1139-0001 can hinder public comment and remove advisory committees

Comments on these proposed regulatory changes are due TODAY, September 10, 2026!

Go here to leave a comment with the Nuclear Regulatory Commission: https://www.regulations.gov/document/NRC-2025-1139-0001

Below is what Sharon and I just posted a few minutes ago. (Comment Tracking Number: mtv-tp3p-l7uv )


The NRC describes this proposed rulemaking as non-controversial and plans to move directly to a final rule unless it receives " ... significant adverse comments ... ". (The final version of the rule has specific changes to 10 CFR, and is available here: https://www.regulations.gov/document/NRC-2025-1139-0002.)

It is generous to presume most of these changes are merely administrative. However, the wording of the rule doesn't make it clear what the actual impact of the changes will be, since, for example, "substantive" is an ambiguous term.

The final version of the rule has a long and detailed section devoted to "advisory committees" which is aimed at reducing the number and cost of such committees. What will the practical impact of such a change be for society, or the public, and for impacted individuals? The rulemaking does not make any of that clear, and leaves us with vague statements such as:

" ... changes are administrative and procedural in nature and do not introduce new substantive obligations for NRC licensees or other stakeholders ..."

" ... revisions affect terminology, chartering requirements, membership appointment processes, and reporting obligations ... "

Which advisory committees and advisors will be eliminated? Will the NRC be getting less input from industry organizations, such as the Nuclear Energy Institute (NEI), or from companies seeking nuclear licenses, or from experts in the field who do not have a vested interest? Numerous scientific studies support more-restrictive radiation standards — not less. Will support for that position be the first to be eliminated?

Similarly, the rulemaking changes the requirements for additional public comment periods after a rule is announced, because such comment periods were not actually required in some cases. (The NRC had made the requirement for these additional comment periods explicit in an 1985 amendment to 10 CFR, part 2.) The NRC explains the proposed reversal of that amendment by quoting the 1985 final rule which says: "The Commission normally provides for ... comment on policy statements and interpretive rules, and will continue to do so in the future."

However, it is not clear whether the entire 1985 amendment is being eliminated and therefore whether this vague reassurance has any weight. In addition, the underlying reasons for additional public comment periods are ignored in the new rule. How will making the extra comment period optional allow for exposing: " ... any errors or oversights that occurred in the formulation of the rule ... ” which was the intent or the 1985 change to 10 CFR part 2?

There is no discussion in the proposed rule changes for what might be lost by making current requirements for public comment periods discretionary. If the public had the opportunity to comment prior to this new rule — why shouldn't the rule specify that the NRC should give the public at least as much opportunity to comment as is currently available?

This change should be completely discarded.

Ace & Sharon Hoffman, Carlsbad, California USA



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Sunday, September 6, 2026

Comments due Sept. 10! NRC-2025-1237-0001 weakens radiation protection for caregivers; preempts tighter state regs...

Post comments here: https://www.regulations.gov/document/NRC-2025-1237-0001

(Here's what Sharon and I just submitted, NRC Comment Tracking Number: mtq-m7u1-38b4)

September 6, 2026

We oppose implementation of the changes in NRC-2025-1237-0001 which impact radiation standards for medical use of radioactive materials. We specifically ask the NRC to consider the comments and questions raised by the American College of Occupational and Environmental Medicine (ACOEM). (The ACOEM describes its members as "... the physicians who conduct medical surveillance of radiation workers, evaluate individuals after suspected overexposures, counsel patients and their families about radiation risk, and perform the causation analyses on which occupational illness compensation depends.")

The ACOEM states that: " ... the proposed rules as a whole reduces protection for workers, for caregivers, and for members of the public on a scientific record that the NRC itself describes as supporting the opposite conclusion." If this statement was the only concern raised about this proposed change to regulations, it should be sufficient to prevent the new rules from being implemented until the ACOEM's concerns are addressed.

The ACOEM raises important concerns about radiation exposure standards for caregivers such as a parent or spouse (quotes from the ACOEM's comment).

  • "Caregivers receive no radiation worker training, no individual monitoring, no medical surveillance, and no periodic dose accounting." Yet, the proposed standards would make it possible for caregivers to receive even more radiation than professionals whose dose limits are restricted by occupational safety standards.
  • There is no clarity about the total radiation a caregiver might receive. For example, a caregiver might be exposed to radiation while visiting the patient in a medical facility, and then be exposed to additional radiation while caring for the patient at home.
  • Special radiation limits exist for pregnant workers, but not for caregivers who may be pregnant.
  • The proposed regulations for caregiver exposure do not apply to minors, because minors cannot be designated as caregivers.

The reality is that minors and pregnant women may act as caregivers and they will almost surely be exposed to radiation if a patient is discharged for home care. It is better to acknowledge that people in these categories are particularly vulnerable to radiation and provide them with training, monitoring, and as much protection as possible.

Instead, in addition to not considering these special vulnerabilities for caregivers, the new regulation would reduce reporting requirements for radiation exposure to fetuses. The rational is that a patient might be unaware of her pregnancy when the radiation is administered. This is certainly a valid scenario, but it should not mean that the exposure to the fetus should not be reported. Instead, any child who was exposed to radiation in utero should have that fact as part of their medical record.

Finally, the ACOEM raises important points concerning state radiation protection standards, pointing out that under the proposed regulation: " ... a State could not retain an ALARA requirement, could not retain a ceiling on authorized public dose, and could not retain a lower caregiver dose allowance, even if its own legislature or radiation control program concluded that doing so was warranted for its population."

Throughout the nuclear era there have been efforts to relax the regulations for exposure to radiation. Leslie Groves claimed to have heard that severe radiation poisoning "is a pleasant way to die". In the latest attempts to reduce radiation protection, federal employees and others have promoted the blatantly false assertion that "Hormesis" is a thing — without a shred of evidence, and despite study after study indicating just the opposite: that LNT is a good approximation of radiation risk from exposure, down to the lowest possible measurable amount. BEIR VII confirmed it and literally thousands of individual research projects have contraindicated Hormesis. The physics of how the damage occurs was unknown when man-made radiation was introduced into society , but the fact that damage occurs was well-known before the Manhattan Project.

All ionizing radiation presents risks and should have a corresponding benefit to society. Both authors of this comment have benefited from the high doses of radiation given to diagnose and treat cancer patients (both are cancer survivors) and appreciate the work of organizations such as the ACOEM.

Even small radiation doses to vast numbers of people will cause cancers that need not have happened. Radiation exposure should ALWAYS be kept As Low As Reasonably Achievable (ALARA). These proposed rules will weaken protections for millions of people and will make it difficult for families and their medical providers to assess risks. For these reasons they should be abandoned.

We incorporate by reference and adopt as our own the comments submitted by the ACOEM (https://www.regulations.gov/comment/NRC-2025-1237-0006 tracking #: mtd-chbq-9vvg).

Sharon and Ace Hoffman
Carlsbad California USA (images added to the blog version)



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Thursday, August 27, 2026

The real reason for the sudden push for Small Nuclear Reactors (which will NEVER be "Modular" BTW so SMRs is the wrong name!)

It turns out (see below for one source) that the U.S. Government is "investing" billions of dollars in the future of inefficient, expensive, unreliable, complex, impossible to dispose of, Small Modular Reactors (SMRs). The real question is: WHY???

Face it: The reason has to be something sinister — it just has to be. Nuclear power always has a dark side (recall "Atoms for Peace" was just a cover to get the public to accept atoms for war (nuclear weapons)).

It turns out, it is sinister. Very sinister.

The United States military — currently run by the craziest, laziest, most ignorant ignoramus ever to lead an army — wants Small Modular Reactors. Specifically five of them, to be operating by 2028 (a date undoubtedly picked for political reasons). All five will be located on military bases in America.

How best to describe this? How about: One of the dumbest ideas in the history of dumb nuclear ideas! (And that's saying a lot.)

Meanwhile, dozens of senior military career officers are being unceremoniously fired by the ignoramus. Might this have been one of many things they knew was a bad idea? One would hope so...

At the moment, the "new, tougher, high-test[osterone] army" is "investing" $2.2 BILLION smackeroos (your tax dollars) to get its hands on these five SMRs. And they're to be mobile, as in mobile-Chernobyls (to recoin a phrase from yesteryear's nuclear waste transport plans to ship thousands of highly radioactive used reactor cores to Yucca Mountain, Nevada — which is technically still on the books as the ONLY solution to nuclear waste, even though it's cancelled AND was unworkable to begin with, when it was first proposed about half a century ago.

But the really crazy part is how vulnerable a nuclear reactor would be if it was deployed during a modern war. Think drone swarms, gravity bombs, unmanned tanks, robot soldiers that can outrun a horse, saboteurs, and general mayhem.

The plan, obviously, it to provide power to the fighting forces. So let's looks at that for a moment (which is more than enough time to see how stupid it is):

A typical SMR is expected to produce several hundred megawatts of electricity (roughly a third as much as a "modern" full-scale nuclear reactor (designed in the 1960s and 70s)). These five are called "Microreactors" and range from just one megawatt to twenty megawatts of power.

What would an army need these for? In an emergency would it feed into the civilian grid? (Not likely!) The bases would all still be connected to the local grid, though. Backup power for them, not for us. But of course, it's just a test. They want to take them into battle. Seriously!

To charge their phones? To charge their drones? To charge their robots, unmanned tanks, and the night-lights in their mobile bathrooms?

All of that, perhaps, but mainly, it's assumed, to power their own Data Center, which would also have to be mobile, and would also be a target. Data Centers are already targets during war, and — as with power — distributed data is the most resilient. What will the troops (or the robots) do when their Data Center is blown up (or its power source is disabled, or both, or the connections between the soldier (or robot) in the field and the Date Center are cut)?

One central power unit for what, several thousand troops? That's a target right there — and a rather easy-to-find one, at that! Despite shielding, any operating reactor is constantly emitting large quantities of high-energy rays and particles that are detectable with all sorts of "high-tech" equipment (they would be detectable with your phone if the phone manufacturers felt like adding a micro-sized radiation detector (which would barely raise the price and could provide a much-needed service in an emergency (I'm sure Herman Kahn would have approved!).

Besides that, mobile Data Centers would either be emitting radio-frequency communications to soldiers, robots or drones, or would have fiber-optic or other direct connections. Both Mobile SMRs and Data Centers would, in short, be high-value targets that are easy to find and difficult or impossible to protect.

The following corporations have all finalized contracts with the military to place nuclear reactors on five bases:

Antares Nuclear, Inc. – Deploying at Fort Bragg, North Carolina.

BWXT Advanced Technologies, LLC – Deploying at Fort Campbell, Kentucky.

General Atomics Electromagnetic Systems – Deploying at Fort Hood, Texas.

Radiant Industries, Inc. – Deploying at Fort Benning, Georgia (under a $750 million slice of the program).

Westinghouse Government Services – Deploying at Fort Drum, New York.

The claimed benefit? Resilient, localized backup.

At military bases in America!!! You would think (and Herman Kahn would certainly agree) that making the ENTIRE US GRID resilient would be a FAR BETTER choice for America — both for "defense" purposes as well as for the well-being of our citizens — and our soldiers.

However, these "microreactors" are supposed to be "mobile" reactors, and they're supposed to be deployed — it's MANDATED to have them OPERATIONAL by September, 2028! As if mandating it can FORCE it to happen. It's August, 2026 already, that's one heck of a time-frame to do the impossible — make a SAFE, RELIABLE nuclear reactor. It's not going to happen. They might build these "test" reactors, but they won't be safe or reliable. They might appear to be sort of reliable in peace, but during war they're vulnerable. And when one fails — even the smallest is expected to be large enough to power 1,000 typical U.S. homes, and most of them could power tens of thousands of homes. When that fails, due to the conditions of war, what's an army to do?

In America (if the fighting were on the home turf) the answer should be simple and obvious: plug into the resilient, reliable, civilian power grid (like they do now). Which would be virtually 100% reliable EVEN DURING WAR if it were powered by small-scale renewable energy.

Wind, wave, solar... with energy-dense batteries for backup.

Could you win a war that way? You can't win it with nuclear, that's for sure, because the "mobile" SMRs wouldn't be very mobile during a war: In all likelihood they would have to be shut down, cooled down, partially dismantled, transported (in pieces), and then re-assembled... while under fire and easy detection by the enemy.

Being on foreign soil isn't any better, since the attack could come from all sides — unless we're fighting together with allies, but we seem to have lost all of those, thanks to actions by the current Administration, such as threatening Greenland, NATO, Canada and even Switzerland, all in the last year, while senselessly and ruthlessly attacking smaller and weaker countries out of pure vanity, whilst committing war crimes all the while. This is not a recipe for peace in the world.

Whenever a nuclear reactor is available as a target for the enemy (see Bennett Ramberg's book on the subject) it must be protected. Nuclear reactors in America are — by law — at least minimally protected with armed guards, guard towers, high fences, lines-of-sight carefully thought out (not much use nowadays what with drone swarms the most likely attack vector)...

(Nuclear spent fuel, which is even more dangerous in many cases because there's so much of it in one place, isn't nearly as well guarded/protected, but harder to attack, thankfully. (Not hard enough to prevent it being a target in a thermonuclear war, for example, or by any terrorist hell-bent on destroying an entire state.)

Reactors being used in war are especially vulnerable targets. Disable ONE nuclear aircraft carrier and you put an entire air wing of our Navy out of action, perhaps sinking the carrier along with poisoning and drowning its crew. And you can do this from hundreds of miles away, if you've got a good "gravity bomb", which are nearly impossible to defend against (or if a friend of yours can give you one for a special occasion). One gravity bomb, well-aimed. And targeting is getting better all the time, all over the world.

Nuclear reactors are a waste of military money. The $2.2 Billion could be used for life-saving vaccines and aid to climate-stricken areas, of course — or it could be used to strengthen the civilian grid near each of the five chosen bases, which would be the BEST way to make the base's energy system RELIABLE EVEN IN TIME OF WAR, and the cheapest as well, for that matter. Or it could be used to install MOBILE solar and wind and non-lithium battery backups, easily carrying enough power to fight a modern war, which are only expected to last a few days (see Herman Kahn again, or listen to the current abomination in the White House, who announced he had "won" a war in three days that is still going on nearly six months later).

Ace Hoffman, Carlsbad, California USA

### Note: Nuclear expert Marvin Resnikoff commented on these reactors recently: "It is more than radioactive waste that is involved. SMR's require highly enriched uranium to operate for 8 years or more without refueling. To operate, reactors require water or sodium. To produce electricity, either PWR's or BWR's, require water as steam to turn a turbine."

The above essay was NOT written with AI (it went through about a dozen human edit passes by two humans instead) but is based on this google AI-based email that was passed around this morning:


https://share.google/aimode/tcH1Jn1mtik2a3Ix3

The U.S. Army is investing up to $2.2 billion over five years to deploy more than 20 commercial nuclear microreactors at domestic military installations. Announced under the Janus Program, this milestone-based initiative partners with the Defense Innovation Unit (DIU) to decouple critical defense infrastructure from potentially vulnerable civilian electrical grids. [1, 2, 3, 4]

Selected Companies and Military Bases

The Army has finalized contracts with five private vendors to build, own, and operate these systems at five initial base locations: [2, 5]

Antares Nuclear, Inc. – Deploying at Fort Bragg, North Carolina.

BWXT Advanced Technologies, LLC – Deploying at Fort Campbell, Kentucky.

General Atomics Electromagnetic Systems – Deploying at Fort Hood, Texas.

Radiant Industries, Inc. – Deploying at Fort Benning, Georgia (under a $750 million slice of the program).

Westinghouse Government Services – Deploying at Fort Drum, New York. [6, 7]

Key Specifications of the Microreactors

Power Output: Each microreactor will generate between 1 and 20 megawatts of electricity, roughly enough to sustain a small town. [7, 8] Grid Independence: The systems will operate on a "private wire" configuration. Bases remain connected to civilian grids, but microreactors act as a resilient, localized backup. [7, 9]

Mobility: Unlike massive conventional nuclear facilities, these factory-built modules are compact enough to be transported via truck, rail, or heavy cargo aircraft. [10, 11]

Timeline: Driven by Executive Order 14299, the Pentagon is legally mandated to have an operational nuclear reactor on a domestic military installation by September 30, 2028. [9, 12]

Why the Military is Going Nuclear

The military wants to replace its reliance on diesel backup generators, which present a logistically vulnerable fuel supply chain during conflicts. Advanced microreactors can operate for years without refueling, providing consistent baseload power capable of running energy-intensive, next-generation weapons systems and command centers. [4, 7, 13]

Furthermore, the U.S. Department of the Army intends for the Janus Program to serve as the "spear tip" for the commercial energy industry. By absorbing the initial financial and regulatory risks, the military is helping private developers scale up mass manufacturing to pave the way for civilian commercial use. [3, 7, 14, 15]

Are you tracking this deployment for its implications on defense technology, or are you more interested in how it affects the broader commercial nuclear energy sector?

[1] https://abcnews.com
[2] https://www.powermag.com
[3] https://www.army.mil
[4] https://www.houstonchronicle.com
[5] https://defensescoop.com
[6] https://www.defenseone.com
[7] https://www.krqe.com
[8] https://www.wsj.com
[9] https://neutronbytes.com
[10] https://www.nytimes.com
[11] https://www.armytimes.com
[12] https://www.bwxt.com
[13] https://www.reuters.com
[14] https://energy.media
[15] https://www.cnbc.com



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Sunday, August 23, 2026

Comments due Aug. 26, 2026 for NRC-2025-1667-0001 which will weaken radioactive transport standards!

Regulations governing the packaging and transportation of radioactive material

Post comments here: https://www.regulations.gov/document/NRC-2025-1667-0001

This is what Sharon and I just posted, with NRC Comment Tracking Number mt6-m1cx-nl1k

August 23, 2026

We strongly oppose proposed changes to 10 CFR part 71 as described in NRC-2025-1667-0001 covering: " ... regulations governing the packaging and transportation of radioactive material ... ". The proposed changes are designed to: " ... support the future domestic deployment of microreactors and other similar new nuclear technologies ... ".

The proposed change would: "... Allow increased radiation level limits, under certain conditions ...", because there may be: "... accelerated timelines that may not allow for extended decay heat or radiation level reduction through post-irradiation cooling ...". So, instead of requiring a design and a transportation plan that meets radiation standards, the new version of Part 71 would reduce radiation protection.

If the proposed changes to Part 71 are implemented, companies that transport radioactive materials will save money. But the public (including unborn generations and people outside the United States) will pay the health and accident costs.

The NRC claims that: " ... the proposed rulemaking would not impact threatened or endangered species or critical habitat ...", but the NRC does not explain why, except to say that: " ... the proposed rule provisions would not have a significant effect on the quality of the human environment ... ". Whether or not the second statement is correct, what happens in the human environment may be totally different from what happens to animals living in a critical habitat such as a river, national forest, etc..

The proposed rule change says that: " ... transportation of radioactive material is one of the safest sectors of hazardous materials logistics ... " because there have been: " ... no recorded fatalities or serious injuries attributable to radiation exposure during transport or as a result of a transportation accident." How would we know? It is virtually impossible to prove that a particular radiation exposure led to a fatality or serious injury. Furthermore, under these regulations, thousands of reactors will be transported around the country, according to the NRC's own estimates. Serious accidents will be risked with each shipment. The totality of risk must be considered.

The NRC admits that fatalities and serious injuries can occur as a result of radiation exposure -- that's exactly why there are radiation standards. But the standards don't guarantee that no living things will die or suffer generational damage -- they only reduce the probabilities for "reference man". Increased levels of radiation exposure should only occur (if at all) after careful consideration of the true costs to the environment, to all humans (not just "reference man"), and to all living things. Alternatives, including wind and solar power, must be properly considered for their lower risk, lower cost, faster deployment, recyclable components, etc..

As a society, we should not accept that multiple generations could suffer genetic harm, for example just because a pregnant woman happens to be stuck in a traffic jam next to a vehicle that is carrying radioactive material that is allowed by the increased radiation limits of this new regulation.

This proposed change is one of many the NRC has published recently. As the NRC itself explained during a recent public meeting, all of these regulations are dependent on each other and the order in which they are approved matters. So, for instance, if this regulation is approved, any changes to 10 CFR Part 20 (Standards for Protection Against Radiation) might have to be revised. There are several ways this intertwined approval process might subvert the requirements of other regulations (both NRC regulations and those from other federal, state, and local agencies). For example, the proposed rule for Reforming and Modernizing the Radiation Protection Framework (https://www.regulations.gov/document/NRC-2025-1140-0001) has 576 comments as of August 23, 2026 whereas the proposed change to Part 71 (this regulation) has only 3 comments. However if the changes to Part 71 are approved first, many of the comments concerning Part 20 could be negated without evaluating the concerns that they raise.

Multiple groups, including other government agencies, have requested extensions of specific comment deadlines. For example, the Conference of Radiation Control Program Directors (CRCPD), requested a 15-day extension of the comment deadline for NRC-2025-1667-0001 (this proposed regulation). The CRCPD describes itself as the national nonprofit organization representing the radiation control programs of all 50 states, the District of Columbia, and the U.S. territories, and provided a clear explanation of why a small extension for this particular deadline would be useful.

We urge the NRC to reconsider the CRCPD's request for a comment-period extension, and to provide a clear answer concerning how comments on the overlapping proposals for rule changes will be reconciled.

Sharon & Ace Hoffman, Carlsbad California USA



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company



Saturday, August 22, 2026

Comments due Aug. 24, 2026 re NRC-2026-2476-0002 which would reduce safeguards for spent nuclear fuel

Post comments to the NRC by the end of Monday, August 24, 2026: https://www.regulations.gov/document/NRC-2026-2476-0002

Below is what we posted to the NRC; the NRC's Comment Tracking Number for this comment is: mt4-zo1t-cbak

(All quotes are from this proposed rule.)

Re: NRC-2026-2476-0002:

List of Approved Spent Fuel Storage Casks: Holtec International HI-STORM Flood/Wind System, Certificate of Compliance No. 1032, Amendment No. 10

August 22, 2026
We strongly oppose the proposed change described in NRC-2026-2476-0002, which would reduce many safeguards for a specific dry-cask design: Holtec's HI-STORM FW. This proposal would enable the Nuclear Regulatory Commission (NRC) to approve changes to dry-cask requirements for a particular dry-cask design as if there was no potential impact on health and safety.

According to the NRC's summary, the proposed change allows design enhancements that:

" ... accommodates two multipurpose canisters stored vertically ... "

" ... higher allowable per-canister and per-cell heat-load limits ... "

" ... reduction of minimum cooling-time requirements ... " (Specifically, Holtec is requesting that the NRC: "... Reduce MPC–44 minimum cooling time from 3 years to 1 year, and MPC–37P minimum cooling time from 1.6 years to 1 year."

" ... refinement of the missile-impact analysis ... "

Despite all these changes that make it possible for more radiation to escape from the dry-casks and for the casks to degrade more quickly, the NRC contends that " ... there will be no significant change in the types or amounts of any effluent released, no significant increase in the individual or cumulative radiation exposure, and no significant increase in the potential for or consequences from radiological accidents." In reality IF an accident or missile impact happens after one year of cooling versus after three years of cooling, the radiological consequences would be much more severe for the shorter time frame, "hotter" spent fuel.

The rule-making proposal explicitly states that: " ... the NRC considers this action to be non-controversial ... ". Therefore, unless the NRC receives what the agency describes as " ... significant adverse comment ... " the final rule will automatically become effective on October 6, 2026. Please consider THESE comments to be SIGNIFICANT. They represent more than 100 years (combined by the two authors) of study in various related issues, including conversations with three Manhattan Project scientists and dozens, if not hundreds, of nuclear engineers, nuclear physicists, nuclear operators, nuclear plant workers, and professionals in related fields including metallurgy, radiation health physics, engineering, safety analysis, computer simulation, etc. etc.. The views presented, however, are the authors' alone.

The NRC should explain the analysis by which it believes that subjecting a dry-cask to the most intense radiation for an additional one to three years does not increase the potential for, and the potential severity of, radiological accidents, especially after decades of storage during which embrittlement will continuously occur — not might occur, but will occur. Additionally, the NRC should be doing a missile-impact analysis for gravity bombs, not just small-arms impacts from Rocket-Propelled Grenades or drone-impacted explosives, or whatever they think the "most likely attack vector" is. Missile impact requirements should be based on the most damaging attack vectors (while not ignoring the most likely). Sharon and Ace Hoffman

Carlsbad California USA

The above comment was submitted on Aug 22, 2026 and received NRC comment tracking id:

mt4-zo1t-cbak



Contact information for the author of this newsletter:

Ace Hoffman
Carlsbad, California USA
Author, The Code Killers:
An Expose of the Nuclear Industry
Free download: acehoffman.org
Blog: acehoffman.blogspot.com
YouTube: youtube.com/user/AceHoffman
Email: ace [at] acehoffman.org
Founder & Owner, The Animated Software Company